File 028974

Images

Text

From: Moshe Hoffman
Sent: 5/3/2016 4:51:28 PM
To: jeffrey E. [jeevacation@gmail.com]
Subject: Greetings, Thanks, and Top Five Questions for the Social Sciences (which Martin thought may interest you)
Importance: High
Hi Jeff,
Thanks again for making the time for me last time you were in town. And for your introductions to Brockman
and Ehud. I have enjoyed all the conversations and been learning a lot. Hopefully, a book or two will come out
of it eventually as well. Means a lot to me.
Martin mentioned that you are thinking about the top questions in various fields and suggested I might think of
some of my own and pass them on to you.
In case you are interested, I list below the five biggest questions that keep me up at night, motivate my life's
work, and which I think ought to keep up every social scientist. Happy to discuss with you, if and when you are
interested.
-Moshe
1) Where do our moral and political views come from?
The bull shit answers most people accept clearly don't fit the facts; we are not discovering moral truths through
reason, and our political views are not motivated by a desire to achieve the best policy outcomes. So what does
drive these beliefs? And what causes them to have the weird puzzling features they do (e.g., why is a lie of
commission so much worse than a lie of omission)? And what causes them to change over time (e.g., slavery
has been accepted as moral at many points in history)? And differ across cultures (e.g., ISIS versus the U.S.)?
And across person and context (e.g., Trump vs. Bernie supporters)?
More generally, we believe and argue all sorts of crap (your vote can make a difference, Jesus loves you, all
men are created equal). How do these beliefs and arguments work? Are they just random ideas propagated by
an amorphous culture, viruses taking advantage of our own psychology for their benefit, ideas that get us to do
the bidding of their cynical designers?
2) Similarly, where do our tastes come from? Like the art we like? Or the music?
Some tastes are kinda obvious. E.g., we evolved to like young curvy women because they are the most fertile.
And maybe we like art that reminds us of this or of landscapes that are safe or what not. But there are many
aspects of art and music (e.g. modern art, rap) that are not universally pleasing. So what makes people like these
things? What properties do they need to succeed? What role does the artist or the history of the piece play in
what we like? Of course, this question isn't just about art and music, but about all of our peculiar, culturally
specific tastes. How does culture shape our tastes? Is it completely arbitrary or is there some logic to its
influence?
3) Where do our political and economic institutions come from and what causes them to have the peculiar
structure they have?
HOUSE OVERSIGHT 028974
We have one vote per person, but everyone knows dollars can buy more votes, and voters in small states are
better represented than those in large states. Whose interests are being served and why theirs? Many economists
and political scientists have thought about "optimal institutions." And much comparative and historical work
has been done. But there isn't really a good unified theory of how these evolve, what characteristics are likely to
emerge, and in what ways they are or are not optimal.
4) Is there a fundamental theory of human societies that's analogous to Darwin's theory in biology?
Can that framework be as well supported by evidence and as simple as natural selection in biology, that will
likewise explain the interesting things humans do, like, and believe, and the institutions we construct?
Evolutionary psychologists think this doesn't require a new theory, just an investigation of the mind that
biologically evolved on the Savannah. Others like Dawkins have suggested that we are best understood as
machines infested by viruses of the mind which themselves evolve to take advantage of our evolved
psychology. Anthropologists treat culture as just random inputs. Social psychologists document the funny
quirks of our behavior without explaining where they come from. And economists assume people optimize
given their tastes and beliefs but take those as given. Is one of these the right theory, if not is there a more
compelling theory out there? What will it look like? How will we find it? How will we know we have found it,
when we have?
5) How can we fix the perverse incentives in academia?
Researchers get hired and promoted for publishing in top journals. Journal publication is determined by editors
and referees. The best way to publish is to make sure your allies are the referees, cite their work, and don't step
on their toes, and to convince the editor you are famous enough and your work is flashy enough that it will be
well cited and get covered by the nytimes. None of this requires much attention be paid to truth, insight, or
originally.
Arguably, this is why the social sciences are such a mess. Few bother to incorporate evidence or arguments
from outsiders (e.g., social psychologists don't consult historians, economists barely consult social
psychologists). Few bother to question the fundamental assumptions driving their field that make no sense and
go unchecked (e.g., anthropologists treat culture as random inputs). And many researchers end up spending
their whole careers developing and propagating theories that are completely uninteresting to outsiders, or
obviously fallacious to anyone not steeped in their literature.
It is ridiculous that we're still using a system derived from the Society of Letters of the 1500s. Perhaps it's time
for something more like Wikipedia, where information gets aggregated and contributors are incentivizes by
their reputation as competent editors to integrate information from across disciplines? Or something like Reddit,
where quality research is recognized by upvotes that aggregate the opinion of thousands from diverse fields? Or
a certification system of sorts, where papers are certified as having done their statistics right or having
integrated what's known from certain fields, or not citing irrelevant papers for political reasons? Whatever the
solution is, social scientists and engineers who work on this problem will have orders of magnitude more impact
on science than scientists doing primary research in our archaic system.
HOUSE OVERSIGHT 028975